The effect on hospital admission profiles of establishing an emergency department observation ward

Objective: To determine the effect of establishing an emergency department observation ward (OW) on admission numbers, average length of stay (ALOS) f or the entire hospital and overall bed days for conditions commonly treated in the OW. Setting: Sir Charles Gairdner Hospital (SCGH), Perth, a tertiary referral t eaching hospital. Design: Retrospective analysis of routinely collected hospital data for the 10 most common diagnosis-related group (DRG) categories of patients discha rged from the OW for the financial years 1995-96 to 1998-99. Comparison of these data with those for adult patients at the other Perth teaching hospit als over the same period. Main outcome measures: For patients in the 10 most common DRGs: numbers of admissions to the OW compared with other inpatient wards; total number of p atients admitted to the hospital compared with total bed days; ALOS at SCGH compared with other Perth teaching hospitals. Results: Increased admissions to the OW were paralleled by a decrease in ad missions for the same DRG codes to other inpatient wards. ALOS remained app roximately the same from 1995-96 to 1998-99 for patients in the OW (one day ) and other inpatient wards (4.38 to 4.20 days). However, overall ALOS for patients in these DRGs fell by over a third (from 3.97 to 2.59 days) over t his time. The total number of patients in these DRGs treated by the hospita l increased by 19% over the four years, but the total number of bed days fe ll by 23%. By contrast, the ALOS for patients in the same DRGs treated at t he other Perth teaching hospitals rose 8% (from 2.12 to 2.28 days). Conclusion: Establishment of a formal emergency department OW results in th e more efficient management of certain groups of patients, with a decrease in overall hospital bed days and length of stay.